File Download
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1145/3127479.3128609
- Scopus: eid_2-s2.0-85032447790
- WOS: WOS:000414279000008
Supplementary
- Citations:
- Appears in Collections:
Conference Paper: APUS: Fast and Scalable PAXOS on RDMA
Title | APUS: Fast and Scalable PAXOS on RDMA |
---|---|
Authors | |
Keywords | State Machine Replication Fault Tolerance Remote Direct Memory Access Software Reliability |
Issue Date | 2017 |
Publisher | ACM. |
Citation | ACM Symposium on Cloud Computing 2017 (SoCC '17), Santa Clara, CA, 24-27 September 2017. In Proceedings of SoCC ’17, 2017, p. 94-107 How to Cite? |
Abstract | State machine replication (SMR) uses Paxos to enforce the same
inputs for a program (e.g., Redis) replicated on a number of hosts,
tolerating various types of failures. Unfortunately, traditional Paxos
protocols incur prohibitive performance overhead on server programs
due to their high consensus latency on TCP/IP. Worse, the
consensus latency of extant Paxos protocols increases drastically
when more concurrent client connections or hosts are added. This
paper presents APUS, the first RDMA-based Paxos protocol that
aims to be fast and scalable to client connections and hosts. APUS
intercepts inbound socket calls of an unmodified server program,
assigns a total order for all input requests, and uses fast RDMA
primitives to replicate these requests concurrently.
We evaluated APUS on nine widely-used server programs (e.g.,
Redis and MySQL). APUS incurred a mean overhead of 4.3% in
response time and 4.2% in throughput. We integrated APUS with an
SMR system Calvin. Our Calvin-APUS integration was 8.2X faster
than the extant Calvin-ZooKeeper integration. The consensus
latency of APUS outperformed an RDMA-based consensus protocol
by 4.9X. APUS source code and raw results are released on github.
com/hku-systems/apus. |
Persistent Identifier | http://hdl.handle.net/10722/245447 |
ISBN | |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Wang, C | - |
dc.contributor.author | Jiang, J | - |
dc.contributor.author | Chen, X | - |
dc.contributor.author | YI, N | - |
dc.contributor.author | Cui, H | - |
dc.date.accessioned | 2017-09-18T02:10:53Z | - |
dc.date.available | 2017-09-18T02:10:53Z | - |
dc.date.issued | 2017 | - |
dc.identifier.citation | ACM Symposium on Cloud Computing 2017 (SoCC '17), Santa Clara, CA, 24-27 September 2017. In Proceedings of SoCC ’17, 2017, p. 94-107 | - |
dc.identifier.isbn | 978-1-4503-5028-0 | - |
dc.identifier.uri | http://hdl.handle.net/10722/245447 | - |
dc.description.abstract | State machine replication (SMR) uses Paxos to enforce the same inputs for a program (e.g., Redis) replicated on a number of hosts, tolerating various types of failures. Unfortunately, traditional Paxos protocols incur prohibitive performance overhead on server programs due to their high consensus latency on TCP/IP. Worse, the consensus latency of extant Paxos protocols increases drastically when more concurrent client connections or hosts are added. This paper presents APUS, the first RDMA-based Paxos protocol that aims to be fast and scalable to client connections and hosts. APUS intercepts inbound socket calls of an unmodified server program, assigns a total order for all input requests, and uses fast RDMA primitives to replicate these requests concurrently. We evaluated APUS on nine widely-used server programs (e.g., Redis and MySQL). APUS incurred a mean overhead of 4.3% in response time and 4.2% in throughput. We integrated APUS with an SMR system Calvin. Our Calvin-APUS integration was 8.2X faster than the extant Calvin-ZooKeeper integration. The consensus latency of APUS outperformed an RDMA-based consensus protocol by 4.9X. APUS source code and raw results are released on github. com/hku-systems/apus. | - |
dc.language | eng | - |
dc.publisher | ACM. | - |
dc.relation.ispartof | Proceedings of SoCC ’17 | - |
dc.subject | State Machine Replication | - |
dc.subject | Fault Tolerance | - |
dc.subject | Remote Direct Memory Access | - |
dc.subject | Software Reliability | - |
dc.title | APUS: Fast and Scalable PAXOS on RDMA | - |
dc.type | Conference_Paper | - |
dc.identifier.email | Cui, H: heming@hku.hk | - |
dc.identifier.authority | Cui, H=rp02008 | - |
dc.description.nature | published_or_final_version | - |
dc.identifier.doi | 10.1145/3127479.3128609 | - |
dc.identifier.scopus | eid_2-s2.0-85032447790 | - |
dc.identifier.hkuros | 276666 | - |
dc.identifier.spage | 94 | - |
dc.identifier.epage | 107 | - |
dc.identifier.isi | WOS:000414279000008 | - |
dc.publisher.place | Santa Clara, CA | - |